A photocatalytic reaction of titanium dioxide was reported in Nature in 1972 and became known world-wide as the Honda-Fujishima effect. Since then, research has been conducted on the production of hydrogen and carbon dioxide by the decomposition of water or the decomposition of aqueous solutions of organic matter by means of titanium dioxide under irradiation with light; and today, a technique in which fine particles of titanium dioxide are held in the form of a thin film on tiles or window glass is in the process of being adapted for practical use in the decomposition of environmental contaminants, i.e., tobacco tar and organic matter such as bacteria, toxins produced by bacteria, etc.
Titanium dioxide is a powder-form metal oxide and is used after being dispersed in a solution in the case of decomposition of water or solutions. However, it is desirable that titanium dioxide adhere to window glass and bathroom tiles, or to the surfaces of construction materials, in the form of a uniform thin film even when titanium dioxide is used in particle form. Accordingly, methods such as sol-gel methods, spray pyrolysis methods using titanium acetate, etc., and dip coating methods, etc. have been developed. These techniques are described in "Oyo Butsuri (Applied Physics)", Vol. 64, No. 8, p. 803 (1995), "Kayak to Kogyo (Chemistry and Industry)", Vol. 48, No. 10, p. 1256 (1995), and "Kayak to Kogyo (Chemistry and Industry)", Vol. 49, No. 6, p. 764 (1996). It has been shown that adhering oils and tobacco tar can be decomposed while being irradiated with ultraviolet light using a glass, etc. covered with such titanium dioxide. It is difficult to decompose inorganic matter such as dirt and dust, etc.; however, it has been reported that since such inorganic matter adheres with organic substances such as oils, etc. as a binder, the decomposition of organic matter also tends to prevent the adhesion of inorganic matter.
The principle of the action of fine particles of titanium dioxide on materials such as tiles, etc. is based on the photocatalytic characteristics of titanium dioxide as a semiconductor. If titanium dioxide is irradiated with light which exceeds the band gap energy, e.g., ultraviolet light, then the electrons in the valence electron band are excited and undergo a transition to the conduction band, so that positively charged holes are left in the valence electron band, thus producing electron-hole pairs. These electrons and holes move through the titanium dioxide and reach the surface; and the electrons are supplied to oxygen in the air so that O.sub.2.sup.- (super-oxide anion) is produced and other substances are reduced. The holes not only cause direct oxidative decomposition of organic matter but also oxidize water molecules adhering to the surface so that strongly oxidizing hydroxyl radicals are formed, and other substances are oxidized by the oxidizing power of these hydroxyl radicals. The O.sub.2.sup.- reportedly participates in this oxidation process; however, the detailed reaction circuit is still being studied. Thus, organic matter is decomposed into carbon dioxide and water by electron-hole pairs excited by light.
In this research, instances have been found in which the electrons and holes re-couple and disappear prior to the oxidation-reduction of external substances in cases where titanium oxide is used alone. Accordingly, it has been indicated that there are limits to the photocatalytic efficiency of titanium dioxide. The ordinary state of titanium dioxide is a powdered state; considering a single particle of titanium dioxide, there are countless lattice defects such as point defects and plane defects, etc. in the surface and interior portions of the particle. When electrons and holes excited in titanium dioxide by ultraviolet light encounter lattice defects in the process of movement, these electrons and holes are captured by the lattice defects and caused to re-couple. In some cases, furthermore, even if the electrons and holes are able to move to the surface, the electrons and holes re-couple when they approach each other. In order to ameliorate such problems, it is necessary to develop techniques for manufacturing titanium dioxide which is free of lattice defects, and techniques for separating electrons and holes at the surface. In regard to the former techniques, improvements have been made in crystal growth techniques; however, since these techniques have no direct connection with the present invention, a detailed description will be omitted here.
In regard to techniques for separating electrons and holes at the surface, a photocatalyst has been proposed in which an electrode which collects excited electrons is formed on the surface of titanium dioxide, so that holes are separated and collected on the surface of the titanium dioxide, while electrons are separated and collected on the surface of the metal electrode. If this approach is used, electrons can be efficiently collected on the surface of the metal electrode, and holes and electrons can be separated; accordingly, the probability of re-coupling would appear to be lowered. Photocatalysts of this type are referred to as "metal-supporting photocatalysts", and are manufactured by forming metals conventionally used as catalysts, such as Pt (platinum) and Cu (copper), etc. on the surface of titanium dioxide. The idea here is that if the metals have a catalytic effect even when used alone, then a synergistic effect with the catalytic action of titanium dioxide should be exhibited.
Methods that have been developed for manufacturing such metal-supporting photocatalysts include: photo-deposition methods in which the semiconductor is suspended in an aqueous solution of a metal salt, after which a reducing agent is added and irradiation with light is performed; impregnation methods in which the semiconductor is immersed in an aqueous solution of a metal salt and dried, after which a reduction treatment is performed; chemical deposition methods in which the semiconductor is violently agitated in an aqueous solution of a metal salt and a reducing agent is added; and simultaneous precipitation methods in which an aqueous solution of a metal salt is added to the semiconductor raw material and simultaneous precipitation is performed, after which sintering is performed. Furthermore, other methods which have been developed include: kneading methods in which the semiconductor and a powdered metal are mixed in a mortar; shaking mixing methods in which the semiconductor and a powered metal are placed in a vessel and mixed by shaking using a shaker, etc.; and powdered metal addition methods in which the semiconductor and a powdered metal are separately added to reaction product solutions and are then suspended and mixed.
The present inventors successively investigated these methods but were unable to form micron-sized fine metal particles that have a particle diameter of 0.1 microns or greater on the surface of titanium dioxide. In other words, the inventors reached the conclusion that the formation of nano-scale ultra-fine metal particles capable of exhibiting a quantum size effect on the surface of a semiconductor is difficult as long as powdered metals or aqueous solutions of metal salts are used. Furthermore, in such conventional methods, the number of fine metal particles that can be supported on one titanium dioxide particle (i.e., the supported particle density) is limited to the range of a few tens of particles. The reasons for such a small supported particle density are that there are problems in the manufacturing methods used, and at the same time, since the particle diameter of the fine metal particle is large, not many fine metal particles adhere to the surface of one titanium dioxide particle. Thus, according to measurements performed by the present inventors, the photocatalytic efficiency of metal-supporting photocatalysts produced by conventional manufacturing methods shows only about a two- to four-fold reinforcement compared to the photocatalytic efficiency of titanium dioxide used alone.
The present inventors made a theoretical investigation, with reference to FIG. 18, concerning the problem of why the photocatalytic efficiency is not greatly reinforced by micron-sized fine metal particles.
In order to achieve efficient incorporation of the electrons generated in titanium dioxide into a metal electrode, it is desirable that the electron transition barrier at the interface between the titanium dioxide and the metal be as small as possible. However, in cases where the particle diameter of the fine metal particles is micron-sized (approximately 0.1 microns or greater), the electron state has roughly the same band structure as a large solid crystal (bulk crystal). In other words, a structure is obtained in which the valence electron band and the conduction band are distinctly formed, with the bands being separated by a fixed band gap, and with free electrons being densely packed in order from the bottom to the uppermost Fermi level in the conduction band. Meanwhile, since titanium oxide is a bulk crystal, the electron state naturally adopts a band structure. In such a band structure, the energy levels constituting the bands are densely arranged more or less continuously, and wave functions corresponding to the respective levels are sharply localized within the substance. In other words, since the wave functions do not extend beyond the substance, the probability that the electrons residing at these levels will be released to the outside of the substance is fairly small.
When, in this state, titanium dioxide is irradiated with ultraviolet light so that electrons are excited into the conduction band so as to cause electron-hole pairs to be formed, in order for these electrons to reduce an external substance A and produce super-oxide anions, it is necessary that the electrons quickly move from the titanium dioxide into the metal, and that the electrons further move from the metal into external substance A outside the metal. However, as described above, since the fine metal particles are micron-sized particles, not only does the electron state adopt a band structure similar to that of a large crystal, but the wave functions also have a structure which is sharply localized within the fine metal particles. Accordingly, since it is not easy for the electrons which have claimed to the conduction band of the titanium dioxide to enter the wave functions of the metal, it is likewise not easy for the electrons to move to the conduction band of the metal. Furthermore, even if the electrons somehow manage to move into the metal, it is similarly difficult for the electrons to move from the metal into an external substance; and in most cases, therefore, before the electrons can leave the metal, the electrons quickly drop to the Fermi level E.sub.F which is in the conduction band of the metal, so that the chance of reacting with the external substance is further diminished.
More specifically, in cases where the level density of the conduction band is high as in the band structure of a bulk crystal, the time required for electrons to drop to the Fermi level (i.e., the relaxation time) is extremely short; and this, together with the localization of the wave functions, prevents the movement of electrons to the outside. In other words, in the case of micron-sized particles, since it is difficult for electrons to move to the outside, electrons are accumulated to an excessive degree inside the metal, so that the movement of electrons from the titanium dioxide into the metal is conversely prevented by the repelling electric field. In the final analysis, it may be concluded that in cases where the particle diameter of the fine metal particles is in the micron size region, electrons remain inside the titanium dioxide or fine metal particles as a result of the energy band structure and localization of the wave functions, so that the probability of electrons to be released to the outside of the metal is reduced to a small value. At the same time, in the case of micron-sized fine metal particles, the number of fine metal particles that can be supported on one titanium dioxide particle is limited to several tens of particles; these are the reasons why the photocatalytic efficiency of such metal-supporting photocatalysts is limited.
Photocatalysts have a decomposing effect on environmental contaminants; and an idea of endowing such photocatalysts with an adsorbing effect has also appeared. Substances that provide such an adsorbing effect include porous materials such as active carbon, active carbon fibers and zeolite, etc. The active carbon fiber is shown in FIG. 19, and in the surface of this fiber, countless pores with a diameter of approximately 0.5 nm, i.e., so-called "micropores", are opened. Environmental contaminants such as organic substances, etc. are adsorbed in these micropores. Since the active carbon fibers can be worked into a variety of shapes, they are widely used in water cleaners and air cleaners.
If such active carbon fibers are used as a base material and caused to hold a photocatalyst, then it should be possible for the active carbon fibers to adsorb environmental contaminants so that the photocatalyst can decompose these environmental contaminants. A deodorizing device in which a photocatalyst is held on active carbon is described in Japanese Patent No. 2574840. FIG. 20 is a conceptual diagram of a photocatalyst in which anatase type titanium dioxide is held on active carbon fibers. If all of the organic matter adsorbed in the micropores were to be decomposed by the photocatalyst, then a adsorbing/decomposing power with an efficiency of 100% would be obtained. However, as described above, since there are limits to the decomposing power of titanium dioxide used alone, some organic matter remains in the micropores. Consequently, the adsorbing power of the active carbon fibers gradually drops until at some point only the decomposing power of the anatase type titanium dioxide remains. Thus, it has been found that the initially expected effect cannot be obtained. The main reason for this is the above-described limit on the photocatalytic efficiency of anatase type titanium dioxide, while a break-through improvement in photocatalytic efficiency is hoped for.
Titanium dioxide includes anatase type titanium dioxide and rutile type titanium dioxide according to differences in crystal structure. Of these two types, rutile type titanium dioxide has a more stable structure. When heated to approximately 600.degree. C. or higher, all anatase type titanium dioxide undergoes a phase transition to a rutile type, and this rutile type remains at low temperatures following cooling. Even at temperatures below 600.degree. C., a portion of such anatase type titanium dioxide is converted into rutile type titanium dioxide. Accordingly, rutile type titanium dioxide can be mass-produced more inexpensively than anatase type titanium dioxide. Conventionally, however, all of the titanium dioxide used as a photocatalyst has been anatase type titanium dioxide, and inexpensive rutile type titanium dioxide has not been used at all. The reason for this may be understood from the band structure.
The band structure of rutile type titanium dioxide is shown in FIG. 21. The gap energy thereof is 3.05 eV. Electrons excited to the conduction band by ultraviolet light reach the bottom of the conduction band while expending a portion of their energy due to relaxation. Since the oxygen potential which is the reduction potential is positioned at 3.13 eV, the climbing of electrons from the bottom of the conduction band to the oxygen potential requires external energy and tends not to happen spontaneously. Accordingly, in the case of rutile type titanium dioxide, it is difficult to form super-oxide anions.
Meanwhile, FIG. 22 shows the band structure of anatase type titanium dioxide. Here, the gap energy thereof is 3.20 eV, so that even if the excited electrons drop to the bottom of the conduction band following excitation by ultraviolet light, oxygen at 3.13 eV can be sufficiently reduced. Thus, anatase type titanium dioxide has the capacity to produce super-oxide anions. Accordingly, in conventional techniques, it has been necessary to use expensive anatase type titanium dioxide as a photocatalyst.
Accordingly, the first object of the present invention is to find a method that makes it possible to achieve a great improvement in the photocatalytic efficiency of titanium dioxide. The second object of the present invention is to find a method which makes it possible for rutile type titanium dioxide, which as not been utilizable as a photocatalyst in the past despite its low cost, to be utilized as a photocatalyst. The third object of the present invention is to provide an effective means of cleaning the environment by causing fine particles of a reinforcing photocatalyst to be held on various types of base materials. The fourth object of the present invention is to realize a photocatalyst which causes an effective continuation of the adsorbing power of active carbon fibers, etc., thus achieving an extended useful life of the adsorption/decomposition cycle and especially to realize a practically effective means of cleaning the environment.